Updated app.py and added examples

.gitattributes

@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/basset.jpg filter=lfs diff=lfs merge=lfs -text
+examples/cat.jpg filter=lfs diff=lfs merge=lfs -text
+examples/dog.jpg filter=lfs diff=lfs merge=lfs -text
+examples/dunno.jpg filter=lfs diff=lfs merge=lfs -text

app.py

@@ -27,37 +27,47 @@ def predict(pickup, dropoff, trip_distance):
     return "The predicted duration is %.4f minutes." % duration
 
 with gr.Blocks() as demo:
-    gr.Markdown("Predict Taxi Duration or Classify dog breeds using this demo")
+    gr.Markdown("""
+    This demo is a simple example of how to use Gradio to create a web interface for your machine learning models.
+
+    Models used in this demo are very simple and are not meant to perform well. The goal is to show how to use Gradio with a simple model.
+    """)
+    
+    gr.Markdown("Predict Taxi Duration or Classify dog vs cat using this demo")
 
     with gr.Tab("Predict Taxi Duration"):
         with gr.Row():
-            pickup = gr.Dropdown(
-                choices=list(zone_lookup["borough_zone"]), 
-                label="Pickup Location", 
-                info="The location where the passenger(s) were picked up", 
-                value=lambda: random.choice(zone_lookup["borough_zone"])
-            )
-
-            dropoff = gr.Dropdown(
-                choices=list(zone_lookup["borough_zone"]), 
-                label="Dropoff Location", 
-                info="The location where the passenger(s) were dropped off", 
-                value=lambda: random.choice(zone_lookup["borough_zone"])
-            )
-
-        trip_distance = gr.Slider(
-            minimum=0.0, 
-            maximum=100.0,
-            step=0.1,
-            label="Trip Distance",
-            info="The trip distance in miles calculated by the taximeter",
-            value=lambda: random.uniform(0.0, 100.0)
-        )
-        with gr.Column():
-            output = gr.Textbox(label="Output Box")
-            predict_btn = gr.Button("Predict")
+            with gr.Column():
+                with gr.Row():
+                    pickup = gr.Dropdown(
+                        choices=list(zone_lookup["borough_zone"]), 
+                        label="Pickup Location", 
+                        info="The location where the passenger(s) were picked up", 
+                        value=lambda: random.choice(zone_lookup["borough_zone"])
+                    )
+
+                    dropoff = gr.Dropdown(
+                        choices=list(zone_lookup["borough_zone"]), 
+                        label="Dropoff Location", 
+                        info="The location where the passenger(s) were dropped off", 
+                        value=lambda: random.choice(zone_lookup["borough_zone"])
+                    )
+
+                trip_distance = gr.Slider(
+                    minimum=0.0, 
+                    maximum=100.0,
+                    step=0.1,
+                    label="Trip Distance",
+                    info="The trip distance in miles calculated by the taximeter",
+                    value=lambda: random.uniform(0.0, 100.0)
+                )
+            with gr.Column():
+                output = gr.Textbox(label="Output Box")
+                predict_btn = gr.Button("Predict")
+
+        examples = gr.Examples([["Queens - Bellerose", "Bronx - Schuylerville/Edgewater Park", 25], ["Bronx - Norwood", "rooklyn - Sunset Park West", 55]], inputs=[pickup, dropoff, trip_distance])
     
-    with gr.Tab("Classify Dog Breed"):
+    with gr.Tab("Classify Dog vs Cat"):
         def is_cat(x): return x[0].isupper()
 
         learn = load_learner('./models/model.pkl')
@@ -68,14 +78,15 @@ with gr.Blocks() as demo:
             pred, idx, probs = learn.predict(img)
             return dict(zip(categories, map(float,probs)))
 
-        image = gr.inputs.Image(shape=(192, 192))
-        label = gr.outputs.Label()
-        examples = ['dog.jpg', 'cat.jpg', 'dunno.jpg']
+        with gr.Row():
+            image = gr.inputs.Image(shape=(192, 192))
+            label = gr.outputs.Label()
+
+        examples = gr.Examples(['./examples/dog.jpg', './examples/cat.jpg', './examples/dunno.jpg', './examples/basset.jpg'], inputs=[image])
         
         classify_btn = gr.Button("Predict")
 
-
     predict_btn.click(fn=predict, inputs=[pickup, dropoff, trip_distance], outputs=output, api_name="predict-duration")       
     classify_btn.click(fn=classify_image, inputs=image, outputs=label, api_name="classify-dog-breed")
 
-demo.launch()
+demo.launch(share=True)